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

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

Neuron 421 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.2772. Min Act: -0.1700

Data Index: 952335 (C4 (Web Text))

Max Activating Token Index: 190

Text #1

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.2772. Min Act: -0.1700

Data Index: 1100102 (C4 (Web Text))

Max Activating Token Index: 677

Text #2

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.2147. Min Act: -0.1700

Data Index: 608016 (C4 (Web Text))

Max Activating Token Index: 336

Text #3

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.1834. Min Act: -0.1700

Data Index: 1116295 (C4 (Web Text))

Max Activating Token Index: 148

Text #4

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.1209. Min Act: -0.1700

Data Index: 223567 (C4 (Web Text))

Max Activating Token Index: 333

Text #5

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.1209. Min Act: -0.1700

Data Index: 1038656 (C4 (Web Text))

Max Activating Token Index: 769

Text #6

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.0897. Min Act: -0.1700

Data Index: 1343368 (C4 (Web Text))

Max Activating Token Index: 304

Text #7

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.0584. Min Act: -0.1700

Data Index: 496631 (C4 (Web Text))

Max Activating Token Index: 470

Text #8

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.0272. Min Act: -0.1700

Data Index: 119864 (C4 (Web Text))

Max Activating Token Index: 377

Text #9

Max Range: 5.2772. Min Range: -5.2772

Max Act: 5.0272. Min Act: -0.1700

Data Index: 152924 (C4 (Web Text))

Max Activating Token Index: 820

Text #10

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9959. Min Act: -0.1700

Data Index: 1065629 (C4 (Web Text))

Max Activating Token Index: 33

Text #11

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9959. Min Act: -0.1700

Data Index: 141751 (C4 (Web Text))

Max Activating Token Index: 385

Text #12

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9334. Min Act: -0.1700

Data Index: 699545 (C4 (Web Text))

Max Activating Token Index: 478

Text #13

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9334. Min Act: -0.1700

Data Index: 27408 (C4 (Web Text))

Max Activating Token Index: 1000

Text #14

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9647. Min Act: -0.1700

Data Index: 194137 (C4 (Web Text))

Max Activating Token Index: 125

Text #15

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9647. Min Act: -0.1700

Data Index: 416318 (C4 (Web Text))

Max Activating Token Index: 658

Text #16

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9647. Min Act: -0.1700

Data Index: 1068601 (C4 (Web Text))

Max Activating Token Index: 369

Text #17

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9334. Min Act: -0.1700

Data Index: 895559 (C4 (Web Text))

Max Activating Token Index: 691

Text #18

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9022. Min Act: -0.1700

Data Index: 567762 (C4 (Web Text))

Max Activating Token Index: 980

Text #19

Max Range: 5.2772. Min Range: -5.2772

Max Act: 4.9647. Min Act: -0.1700

Data Index: 485275 (C4 (Web Text))

Max Activating Token Index: 902