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

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

Neuron 760 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.6264. Min Act: -0.1700

Data Index: 391324 (C4 (Web Text))

Max Activating Token Index: 480

Text #1

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.6264. Min Act: -0.1700

Data Index: 1187924 (C4 (Web Text))

Max Activating Token Index: 7

Text #2

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.4701. Min Act: -0.1700

Data Index: 91803 (C4 (Web Text))

Max Activating Token Index: 547

Text #3

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.4389. Min Act: -0.1700

Data Index: 990236 (C4 (Web Text))

Max Activating Token Index: 272

Text #4

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.2514. Min Act: -0.1700

Data Index: 1342583 (C4 (Web Text))

Max Activating Token Index: 831

Text #5

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.2514. Min Act: -0.1700

Data Index: 762118 (C4 (Web Text))

Max Activating Token Index: 910

Text #6

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.2514. Min Act: -0.1700

Data Index: 1126379 (C4 (Web Text))

Max Activating Token Index: 750

Text #7

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.2514. Min Act: -0.1700

Data Index: 2290 (C4 (Web Text))

Max Activating Token Index: 484

Text #8

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.2514. Min Act: -0.1700

Data Index: 1052644 (C4 (Web Text))

Max Activating Token Index: 24

Text #9

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.1889. Min Act: -0.1700

Data Index: 221397 (C4 (Web Text))

Max Activating Token Index: 954

Text #10

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.1264. Min Act: -0.1700

Data Index: 236919 (C4 (Web Text))

Max Activating Token Index: 337

Text #11

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.1264. Min Act: -0.1700

Data Index: 511900 (C4 (Web Text))

Max Activating Token Index: 28

Text #12

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.1576. Min Act: -0.1700

Data Index: 977859 (C4 (Web Text))

Max Activating Token Index: 235

Text #13

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.1264. Min Act: -0.1700

Data Index: 911234 (C4 (Web Text))

Max Activating Token Index: 658

Text #14

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0639. Min Act: -0.1700

Data Index: 41686 (C4 (Web Text))

Max Activating Token Index: 1005

Text #15

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0951. Min Act: -0.1700

Data Index: 126556 (C4 (Web Text))

Max Activating Token Index: 562

Text #16

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0014. Min Act: -0.1700

Data Index: 427052 (C4 (Web Text))

Max Activating Token Index: 841

Text #17

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0951. Min Act: -0.1700

Data Index: 553743 (C4 (Web Text))

Max Activating Token Index: 136

Text #18

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0639. Min Act: -0.1700

Data Index: 1199721 (C4 (Web Text))

Max Activating Token Index: 894

Text #19

Max Range: 5.6264. Min Range: -5.6264

Max Act: 5.0639. Min Act: -0.1700

Data Index: 298243 (C4 (Web Text))

Max Activating Token Index: 764