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

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

Neuron 148 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.8152. Min Act: -0.1700

Data Index: 98034 (C4 (Web Text))

Max Activating Token Index: 306

Text #1

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.8777. Min Act: -0.1700

Data Index: 914651 (C4 (Web Text))

Max Activating Token Index: 48

Text #2

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.8152. Min Act: -0.1700

Data Index: 408468 (C4 (Web Text))

Max Activating Token Index: 638

Text #3

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.7839. Min Act: -0.1700

Data Index: 509110 (C4 (Web Text))

Max Activating Token Index: 657

Text #4

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.6902. Min Act: -0.1700

Data Index: 496051 (C4 (Web Text))

Max Activating Token Index: 61

Text #5

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.6902. Min Act: -0.1700

Data Index: 23830 (C4 (Web Text))

Max Activating Token Index: 60

Text #6

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.6277. Min Act: -0.1700

Data Index: 640008 (C4 (Web Text))

Max Activating Token Index: 16

Text #7

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5652. Min Act: -0.1700

Data Index: 1118403 (C4 (Web Text))

Max Activating Token Index: 1005

Text #8

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.6277. Min Act: -0.1700

Data Index: 1029104 (C4 (Web Text))

Max Activating Token Index: 190

Text #9

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5964. Min Act: -0.1700

Data Index: 750020 (C4 (Web Text))

Max Activating Token Index: 535

Text #10

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5652. Min Act: -0.1700

Data Index: 1038851 (C4 (Web Text))

Max Activating Token Index: 915

Text #11

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5964. Min Act: -0.1700

Data Index: 214153 (C4 (Web Text))

Max Activating Token Index: 669

Text #12

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5964. Min Act: -0.1700

Data Index: 1165544 (C4 (Web Text))

Max Activating Token Index: 356

Text #13

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5652. Min Act: -0.1700

Data Index: 634378 (C4 (Web Text))

Max Activating Token Index: 433

Text #14

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5652. Min Act: -0.1700

Data Index: 285252 (C4 (Web Text))

Max Activating Token Index: 911

Text #15

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5339. Min Act: -0.1700

Data Index: 383433 (C4 (Web Text))

Max Activating Token Index: 504

Text #16

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5652. Min Act: -0.1700

Data Index: 767924 (C4 (Web Text))

Max Activating Token Index: 766

Text #17

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5339. Min Act: -0.1700

Data Index: 20865 (C4 (Web Text))

Max Activating Token Index: 1019

Text #18

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5339. Min Act: -0.1700

Data Index: 747429 (C4 (Web Text))

Max Activating Token Index: 458

Text #19

Max Range: 5.8777. Min Range: -5.8777

Max Act: 5.5339. Min Act: -0.1700

Data Index: 388041 (C4 (Web Text))

Max Activating Token Index: 343