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

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

Neuron 1100 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: 5.9689. Min Range: -5.9689

Max Act: 5.9689. Min Act: -0.1698

Data Index: 97303 (C4 (Web Text))

Max Activating Token Index: 145

Text #1

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.9377. Min Act: -0.1700

Data Index: 1195346 (C4 (Web Text))

Max Activating Token Index: 298

Text #2

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.6252. Min Act: -0.1700

Data Index: 376777 (C4 (Web Text))

Max Activating Token Index: 938

Text #3

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.6564. Min Act: -0.1700

Data Index: 297807 (C4 (Web Text))

Max Activating Token Index: 172

Text #4

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.6564. Min Act: -0.1700

Data Index: 720287 (C4 (Web Text))

Max Activating Token Index: 944

Text #5

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5939. Min Act: -0.1700

Data Index: 412468 (C4 (Web Text))

Max Activating Token Index: 188

Text #6

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5939. Min Act: -0.1699

Data Index: 92257 (C4 (Web Text))

Max Activating Token Index: 149

Text #7

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5314. Min Act: -0.1700

Data Index: 1005115 (C4 (Web Text))

Max Activating Token Index: 1003

Text #8

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5314. Min Act: -0.1694

Data Index: 596646 (C4 (Web Text))

Max Activating Token Index: 805

Text #9

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5002. Min Act: -0.1699

Data Index: 1050507 (C4 (Web Text))

Max Activating Token Index: 315

Text #10

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4689. Min Act: -0.1700

Data Index: 1219356 (C4 (Web Text))

Max Activating Token Index: 149

Text #11

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.5002. Min Act: -0.1700

Data Index: 906786 (C4 (Web Text))

Max Activating Token Index: 825

Text #12

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4377. Min Act: -0.1698

Data Index: 1054278 (C4 (Web Text))

Max Activating Token Index: 74

Text #13

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4377. Min Act: -0.1700

Data Index: 755537 (C4 (Web Text))

Max Activating Token Index: 894

Text #14

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4377. Min Act: -0.1700

Data Index: 354341 (C4 (Web Text))

Max Activating Token Index: 426

Text #15

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4064. Min Act: -0.1700

Data Index: 860634 (C4 (Web Text))

Max Activating Token Index: 741

Text #16

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4064. Min Act: -0.1699

Data Index: 912936 (C4 (Web Text))

Max Activating Token Index: 333

Text #17

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.4064. Min Act: -0.1699

Data Index: 396209 (C4 (Web Text))

Max Activating Token Index: 505

Text #18

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.3439. Min Act: -0.1700

Data Index: 683986 (C4 (Web Text))

Max Activating Token Index: 389

Text #19

Max Range: 5.9689. Min Range: -5.9689

Max Act: 5.3439. Min Act: -0.1699

Data Index: 903901 (C4 (Web Text))

Max Activating Token Index: 719