Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2001 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.9137. Min Act: -0.1700

Data Index: 1126240 (C4 (Web Text))

Max Activating Token Index: 260

Text #1

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.9137. Min Act: -0.1700

Data Index: 1060784 (C4 (Web Text))

Max Activating Token Index: 991

Text #2

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.8199. Min Act: -0.1700

Data Index: 1122951 (C4 (Web Text))

Max Activating Token Index: 623

Text #3

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.8199. Min Act: -0.1700

Data Index: 1234879 (C4 (Web Text))

Max Activating Token Index: 961

Text #4

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7573. Min Act: -0.1700

Data Index: 258067 (C4 (Web Text))

Max Activating Token Index: 543

Text #5

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7573. Min Act: -0.1700

Data Index: 701826 (C4 (Web Text))

Max Activating Token Index: 19

Text #6

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 1285152 (C4 (Web Text))

Max Activating Token Index: 1008

Text #7

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7573. Min Act: -0.1700

Data Index: 1001659 (C4 (Web Text))

Max Activating Token Index: 451

Text #8

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7260. Min Act: -0.1700

Data Index: 890244 (C4 (Web Text))

Max Activating Token Index: 430

Text #9

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.8199. Min Act: -0.1700

Data Index: 594840 (C4 (Web Text))

Max Activating Token Index: 58

Text #10

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7573. Min Act: -0.1700

Data Index: 1224916 (C4 (Web Text))

Max Activating Token Index: 246

Text #11

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7573. Min Act: -0.1700

Data Index: 96650 (C4 (Web Text))

Max Activating Token Index: 818

Text #12

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 184848 (C4 (Web Text))

Max Activating Token Index: 104

Text #13

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.7260. Min Act: -0.1700

Data Index: 818724 (C4 (Web Text))

Max Activating Token Index: 521

Text #14

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.9137. Min Act: -0.1700

Data Index: 991050 (C4 (Web Text))

Max Activating Token Index: 814

Text #15

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 101995 (C4 (Web Text))

Max Activating Token Index: 624

Text #16

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 799627 (C4 (Web Text))

Max Activating Token Index: 935

Text #17

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 1349982 (C4 (Web Text))

Max Activating Token Index: 480

Text #18

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6947. Min Act: -0.1700

Data Index: 1206298 (C4 (Web Text))

Max Activating Token Index: 78

Text #19

Max Range: 3.9137. Min Range: -3.9137

Max Act: 3.6634. Min Act: -0.1700

Data Index: 121930 (C4 (Web Text))

Max Activating Token Index: 377