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

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

Neuron 2003 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: 7.2678. Min Range: -7.2678

Max Act: 7.2678. Min Act: -0.1700

Data Index: 1142715 (C4 (Web Text))

Max Activating Token Index: 788

Text #1

Max Range: 7.2678. Min Range: -7.2678

Max Act: 7.2053. Min Act: -0.1700

Data Index: 307708 (C4 (Web Text))

Max Activating Token Index: 80

Text #2

Max Range: 7.2678. Min Range: -7.2678

Max Act: 7.1428. Min Act: -0.1700

Data Index: 700590 (C4 (Web Text))

Max Activating Token Index: 92

Text #3

Max Range: 7.2678. Min Range: -7.2678

Max Act: 7.0803. Min Act: -0.1700

Data Index: 679753 (C4 (Web Text))

Max Activating Token Index: 378

Text #4

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.8928. Min Act: -0.1700

Data Index: 1005360 (C4 (Web Text))

Max Activating Token Index: 174

Text #5

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.8303. Min Act: -0.1700

Data Index: 351253 (C4 (Web Text))

Max Activating Token Index: 112

Text #6

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.8928. Min Act: -0.1700

Data Index: 913551 (C4 (Web Text))

Max Activating Token Index: 144

Text #7

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.8303. Min Act: -0.1700

Data Index: 1241010 (C4 (Web Text))

Max Activating Token Index: 864

Text #8

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.7990. Min Act: -0.1700

Data Index: 1346988 (C4 (Web Text))

Max Activating Token Index: 705

Text #9

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6740. Min Act: -0.1700

Data Index: 854087 (C4 (Web Text))

Max Activating Token Index: 357

Text #10

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6740. Min Act: -0.1700

Data Index: 1285039 (C4 (Web Text))

Max Activating Token Index: 158

Text #11

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6428. Min Act: -0.1700

Data Index: 1008124 (C4 (Web Text))

Max Activating Token Index: 814

Text #12

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.5803. Min Act: -0.1700

Data Index: 149940 (C4 (Web Text))

Max Activating Token Index: 702

Text #13

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6428. Min Act: -0.1700

Data Index: 543210 (C4 (Web Text))

Max Activating Token Index: 553

Text #14

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6428. Min Act: -0.1700

Data Index: 610269 (C4 (Web Text))

Max Activating Token Index: 85

Text #15

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6428. Min Act: -0.1700

Data Index: 217868 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.6428. Min Act: -0.1700

Data Index: 761825 (C4 (Web Text))

Max Activating Token Index: 109

Text #17

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.5803. Min Act: -0.1700

Data Index: 77029 (C4 (Web Text))

Max Activating Token Index: 676

Text #18

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.5178. Min Act: -0.1700

Data Index: 1474431 (Python Code)

Max Activating Token Index: 342

Text #19

Max Range: 7.2678. Min Range: -7.2678

Max Act: 6.5178. Min Act: -0.1700

Data Index: 575528 (C4 (Web Text))

Max Activating Token Index: 913