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

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

Neuron 1065 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: 7.1614. Min Range: -7.1614

Max Act: 7.1614. Min Act: -0.1700

Data Index: 286572 (C4 (Web Text))

Max Activating Token Index: 797

Text #1

Max Range: 7.1614. Min Range: -7.1614

Max Act: 7.1614. Min Act: -0.1700

Data Index: 229740 (C4 (Web Text))

Max Activating Token Index: 254

Text #2

Max Range: 7.1614. Min Range: -7.1614

Max Act: 7.1301. Min Act: -0.1700

Data Index: 91703 (C4 (Web Text))

Max Activating Token Index: 250

Text #3

Max Range: 7.1614. Min Range: -7.1614

Max Act: 7.0051. Min Act: -0.1700

Data Index: 209386 (C4 (Web Text))

Max Activating Token Index: 353

Text #4

Max Range: 7.1614. Min Range: -7.1614

Max Act: 7.0051. Min Act: -0.1700

Data Index: 1060149 (C4 (Web Text))

Max Activating Token Index: 183

Text #5

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.9114. Min Act: -0.1700

Data Index: 15013 (C4 (Web Text))

Max Activating Token Index: 920

Text #6

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.9426. Min Act: -0.1700

Data Index: 1342838 (C4 (Web Text))

Max Activating Token Index: 406

Text #7

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.9426. Min Act: -0.1700

Data Index: 485638 (C4 (Web Text))

Max Activating Token Index: 437

Text #8

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8801. Min Act: -0.1700

Data Index: 1161535 (C4 (Web Text))

Max Activating Token Index: 419

Text #9

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8801. Min Act: -0.1700

Data Index: 994182 (C4 (Web Text))

Max Activating Token Index: 445

Text #10

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8489. Min Act: -0.1700

Data Index: 643258 (C4 (Web Text))

Max Activating Token Index: 708

Text #11

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8801. Min Act: -0.1700

Data Index: 687083 (C4 (Web Text))

Max Activating Token Index: 654

Text #12

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8176. Min Act: -0.1700

Data Index: 1151799 (C4 (Web Text))

Max Activating Token Index: 89

Text #13

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.7864. Min Act: -0.1700

Data Index: 1003066 (C4 (Web Text))

Max Activating Token Index: 489

Text #14

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8176. Min Act: -0.1700

Data Index: 755674 (C4 (Web Text))

Max Activating Token Index: 406

Text #15

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.8489. Min Act: -0.1700

Data Index: 550083 (C4 (Web Text))

Max Activating Token Index: 388

Text #16

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.7551. Min Act: -0.1700

Data Index: 1047831 (C4 (Web Text))

Max Activating Token Index: 305

Text #17

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.7551. Min Act: -0.1700

Data Index: 1340305 (C4 (Web Text))

Max Activating Token Index: 858

Text #18

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.7239. Min Act: -0.1700

Data Index: 174548 (C4 (Web Text))

Max Activating Token Index: 751

Text #19

Max Range: 7.1614. Min Range: -7.1614

Max Act: 6.7239. Min Act: -0.1700

Data Index: 447209 (C4 (Web Text))

Max Activating Token Index: 371