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

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

Neuron 203 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: 8.0272. Min Range: -8.0272

Max Act: 7.9334. Min Act: -0.1700

Data Index: 679725 (C4 (Web Text))

Max Activating Token Index: 700

Text #1

Max Range: 8.0272. Min Range: -8.0272

Max Act: 8.0272. Min Act: -0.1700

Data Index: 787961 (C4 (Web Text))

Max Activating Token Index: 481

Text #2

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.8709. Min Act: -0.1700

Data Index: 364148 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.6834. Min Act: -0.1700

Data Index: 605217 (C4 (Web Text))

Max Activating Token Index: 721

Text #4

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.5584. Min Act: -0.1700

Data Index: 1175509 (C4 (Web Text))

Max Activating Token Index: 489

Text #5

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.5584. Min Act: -0.1700

Data Index: 1029587 (C4 (Web Text))

Max Activating Token Index: 429

Text #6

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.1209. Min Act: -0.1700

Data Index: 1199078 (C4 (Web Text))

Max Activating Token Index: 654

Text #7

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.4647. Min Act: -0.1700

Data Index: 282567 (C4 (Web Text))

Max Activating Token Index: 385

Text #8

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.4647. Min Act: -0.1700

Data Index: 1034882 (C4 (Web Text))

Max Activating Token Index: 64

Text #9

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.5272. Min Act: -0.1700

Data Index: 621120 (C4 (Web Text))

Max Activating Token Index: 44

Text #10

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.4022. Min Act: -0.1700

Data Index: 218264 (C4 (Web Text))

Max Activating Token Index: 657

Text #11

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.3709. Min Act: -0.1700

Data Index: 1280606 (C4 (Web Text))

Max Activating Token Index: 373

Text #12

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.3709. Min Act: -0.1700

Data Index: 736080 (C4 (Web Text))

Max Activating Token Index: 119

Text #13

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.4022. Min Act: -0.1700

Data Index: 764463 (C4 (Web Text))

Max Activating Token Index: 488

Text #14

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.5584. Min Act: -0.1700

Data Index: 1290571 (C4 (Web Text))

Max Activating Token Index: 679

Text #15

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.3709. Min Act: -0.1700

Data Index: 20810 (C4 (Web Text))

Max Activating Token Index: 635

Text #16

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.5272. Min Act: -0.1700

Data Index: 179817 (C4 (Web Text))

Max Activating Token Index: 471

Text #17

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.4334. Min Act: -0.1700

Data Index: 725553 (C4 (Web Text))

Max Activating Token Index: 494

Text #18

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.6522. Min Act: -0.1700

Data Index: 509197 (C4 (Web Text))

Max Activating Token Index: 872

Text #19

Max Range: 8.0272. Min Range: -8.0272

Max Act: 7.1834. Min Act: -0.1700

Data Index: 897834 (C4 (Web Text))

Max Activating Token Index: 416